Suicide Re-Event Measure, Adjusting for Suicidal History

• Activity Number: 224
• Type: Topic Contributed
• Date/Time: Monday, August 5, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Mental Health Statistics Section
• Author(s): Brady Stephens*+ and Robert Bossarte
• Companies: US Department of Veteran Affairs and Veterans Administration Medical Center

Abstract

The recently released Veteran Affairs report - Suicide Data Report, 2012 - indicates the annual 12-month re-event rate is between 13% and 16% among Veteran using Veteran Health Administration services. Research suggests that history of a previous suicide event is associated with a higher probability of re-event than many other factors. Comparing over temporal, geographic or other characteristics requires consideration of the underlying suicide event history. Comparisons over time or across administrative units are inherently complicated by the strong associations between history of previous events and re-event risk and changing composition of group membership. A measure currently under evaluation is a weighted suicide re-event measure that discounts the observed re-event percentage using estimates of excess risk associated with previous events. The measure is adjusted based on the inverse of the conditional probability and stratified by event number to generate a cumulative measure of repeat suicide events. The process and underlying assumptions for calculating the proposed metric will be presented and examples of recent analyses presented.